{ "schemaVersion": 3, "id": "ayt-2026-mat", "kind": "exam", "title": { "tr": "2026 AYT · Matematik", "en": "2026 AYT · Mathematics" }, "subtitle": { "tr": "Alan Yeterlilik Testi · Matematik", "en": "Turkish university entrance exam · Mathematics" }, "category": { "tr": "Ulusal sınav", "en": "National exam" }, "description": { "tr": "Yapay zekâ modellerinin 2026 AYT Matematik testindeki doğruluk, puan, maliyet ve token verimliliğini farklı akıl yürütme (reasoning effort) seviyelerinde karşılaştırır.", "en": "Compares AI models on the 2026 AYT Mathematics test across accuracy, score, cost, and token efficiency at different reasoning effort levels." }, "status": "published", "examDate": "2026-06-21", "runId": "2026-07-31T08-46-30Z", "runCompletedAt": "2026-08-12T04:32:58Z", "generatedAt": "2026-08-12T04:32:58Z", "itemCount": 39, "responseCount": 2886, "totalQuestions": 39, "score": { "label": { "tr": "Matematik Puanı", "en": "Mathematics Score" }, "min": 100.0, "max": 217.0, "formula": { "tr": "100 + Matematik neti × 3", "en": "100 + Mathematics net × 3" } }, "netRule": { "tr": "net = doğru − yanlış/4; geçersiz yanıtlar yanlış sayılır.", "en": "net = correct − wrong/4; invalid responses count as wrong." }, "provenance": { "settingsSource": "run", "scoring": { "penaltyDivisor": 4.0, "base": 100.0, "coefficients": { "MAT": 3.0, "FEN": 3.0, "TDE-SB1": 3.0, "SB2": 3.0 } }, "invalidPolicy": "wrong" }, "sections": [ { "code": "MAT", "name": { "tr": "Matematik", "en": "Mathematics" }, "questionCount": 39 } ], "models": [ { "modelId": "openai/gpt-5.6-luna@xhigh", "baseModelId": "openai/gpt-5.6-luna", "modelSlug": "gpt-5-6-luna", "developerId": "openai", "name": "GPT-5.6 Luna (xhigh)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.0372, "known": true, "approximate": false }, "tokens": { "billed": 69598, "known": true, "approximate": false, "prompt": 9152, "completion": 60446, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "xhigh", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 1014.7849, "netPer1kTokens": 0.5424 }, "rank": 1 }, { "modelId": "openai/gpt-5.6-luna@max", "baseModelId": "openai/gpt-5.6-luna", "modelSlug": "gpt-5-6-luna", "developerId": "openai", "name": "GPT-5.6 Luna (max)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.0411, "known": true, "approximate": false }, "tokens": { "billed": 76098, "known": true, "approximate": false, "prompt": 9152, "completion": 66946, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "max", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 918.4915, "netPer1kTokens": 0.4961 }, "rank": 2 }, { "modelId": "inclusionai/ring-2.6-1t@high", "baseModelId": "inclusionai/ring-2.6-1t", "modelSlug": "ring-2-6-1t", "developerId": "inclusion-ai", "name": "Ring-2.6-1T (high)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.0423, "known": true, "approximate": false }, "tokens": { "billed": 77168, "known": true, "approximate": false, "prompt": 10734, "completion": 66434, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 892.435, "netPer1kTokens": 0.4892 }, "rank": 3 }, { "modelId": "inclusionai/ring-2.6-1t@xhigh", "baseModelId": "inclusionai/ring-2.6-1t", "modelSlug": "ring-2-6-1t", "developerId": "inclusion-ai", "name": "Ring-2.6-1T (xhigh)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.0602, "known": true, "approximate": false }, "tokens": { "billed": 105773, "known": true, "approximate": false, "prompt": 10773, "completion": 95000, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "xhigh", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 627.0764, "netPer1kTokens": 0.3569 }, "rank": 4 }, { "modelId": "openai/gpt-5.6-terra@low", "baseModelId": "openai/gpt-5.6-terra", "modelSlug": "gpt-5-6-terra", "developerId": "openai", "name": "GPT-5.6 Terra (low)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.1002, "known": true, "approximate": false }, "tokens": { "billed": 24333, "known": true, "approximate": false, "prompt": 9152, "completion": 15181, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "low", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 376.7465, "netPer1kTokens": 1.5514 }, "rank": 5 }, { "modelId": "openai/gpt-5.6-terra@high", "baseModelId": "openai/gpt-5.6-terra", "modelSlug": "gpt-5-6-terra", "developerId": "openai", "name": "GPT-5.6 Terra (high)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.1147, "known": true, "approximate": false }, "tokens": { "billed": 26749, "known": true, "approximate": false, "prompt": 9152, "completion": 17597, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 329.1194, "netPer1kTokens": 1.4113 }, "rank": 6 }, { "modelId": "openai/gpt-5.6-terra@xhigh", "baseModelId": "openai/gpt-5.6-terra", "modelSlug": "gpt-5-6-terra", "developerId": "openai", "name": "GPT-5.6 Terra (xhigh)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.1408, "known": true, "approximate": false }, "tokens": { "billed": 31086, "known": true, "approximate": false, "prompt": 9152, "completion": 21934, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "xhigh", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 268.1108, "netPer1kTokens": 1.2144 }, "rank": 7 }, { "modelId": "tencent/hy3@high", "baseModelId": "tencent/hy3", "modelSlug": "tencent-hy3", "developerId": "tencent", "name": "Tencent Hy3 (high)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.1496, "known": true, "approximate": true }, "tokens": { "billed": 291328, "known": true, "approximate": false, "prompt": 10756, "completion": 280572, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": { "tr": "Maliyet liste fiyatından hesaplandı; allow_fallbacks nedeniyle 2026-07-31 ölçümünde gerçek fatura hesaplananın ≈1,5 katıydı.", "en": "Cost computed from the list price; with allow_fallbacks, a 2026-07-31 probe measured the actual bill at ≈1.5× the computed value." }, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 252.3396, "netPer1kTokens": 0.1296 }, "rank": 8 }, { "modelId": "meta/muse-glimmer-30b@high", "baseModelId": "meta/muse-glimmer-30b", "modelSlug": "muse-glimmer-30b", "developerId": "meta", "name": "Muse Glimmer 30B (high)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.1588, "known": true, "approximate": false }, "tokens": { "billed": 113764, "known": true, "approximate": false, "prompt": 9319, "completion": 104445, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": { "tr": "Koşu ayarlarındaki 65.536 istenen tavandır; efektif tavan 32.768'dir. Diğer varyantlar 131.072 ile koştu, ancak modelin bağlamı 131.072 olduğundan genel tavan gönderilemezdi. Pinlenen Together endpoint'i ilan etmediği bir sınırla talebi 32.768'de kesiyor. 2026-08-12 ölçümünde bu tavanda hiçbir yanıt kesilmedi.", "en": "The 65,536 shown in run settings is the requested ceiling; the effective ceiling is 32,768. The other variants ran with 131,072, but the run-wide ceiling could not be sent because this model's context is 131,072. The pinned Together endpoint enforces an undeclared limit that caps the request at 32,768. In the 2026-08-12 measurement no response was truncated at this ceiling." }, "settings": { "temperature": 0.0, "maxTokens": 65536, "reasoningEffort": "high", "requestedRoutingPolicy": { "ignore": [ "deepinfra" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 237.7204, "netPer1kTokens": 0.3318 }, "rank": 9 }, { "modelId": "thinkingmachines/inkling@medium", "baseModelId": "thinkingmachines/inkling", "modelSlug": "thinky-inkling", "developerId": "thinking-machines", "name": "Thinky Inkling (medium)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.2546, "known": true, "approximate": false }, "tokens": { "billed": 69985, "known": true, "approximate": false, "prompt": 9464, "completion": 60521, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "medium", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 148.2718, "netPer1kTokens": 0.5394 }, "rank": 10 }, { "modelId": "qwen/qwen3.7-plus", "baseModelId": "qwen/qwen3.7-plus", "modelSlug": "qwen3-7-plus", "developerId": "alibaba-qwen", "name": "Qwen3.7 Plus", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.2587, "known": true, "approximate": false }, "tokens": { "billed": 209436, "known": true, "approximate": false, "prompt": 9734, "completion": 199702, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": null, "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 145.9219, "netPer1kTokens": 0.1802 }, "rank": 11 }, { "modelId": "anthropic/claude-opus-5@low", "baseModelId": "anthropic/claude-opus-5", "modelSlug": "claude-opus-5", "developerId": "anthropic", "name": "Claude Opus 5 (low)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.3064, "known": true, "approximate": false }, "tokens": { "billed": 23735, "known": true, "approximate": false, "prompt": 14351, "completion": 9384, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "low", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 123.205, "netPer1kTokens": 1.5905 }, "rank": 12 }, { "modelId": "openai/gpt-5.6-terra@max", "baseModelId": "openai/gpt-5.6-terra", "modelSlug": "gpt-5-6-terra", "developerId": "openai", "name": "GPT-5.6 Terra (max)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.3304, "known": true, "approximate": false }, "tokens": { "billed": 62695, "known": true, "approximate": false, "prompt": 9152, "completion": 53543, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "max", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 114.2554, "netPer1kTokens": 0.6021 }, "rank": 13 }, { "modelId": "google/gemini-3.6-flash@low", "baseModelId": "google/gemini-3.6-flash", "modelSlug": "gemini-3-6-flash", "developerId": "google", "name": "Gemini 3.6 Flash (low)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.4022, "known": true, "approximate": false }, "tokens": { "billed": 60641, "known": true, "approximate": false, "prompt": 8760, "completion": 51881, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "low", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 93.8588, "netPer1kTokens": 0.6225 }, "rank": 14 }, { "modelId": "anthropic/claude-opus-5@medium", "baseModelId": "anthropic/claude-opus-5", "modelSlug": "claude-opus-5", "developerId": "anthropic", "name": "Claude Opus 5 (medium)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.4677, "known": true, "approximate": false }, "tokens": { "billed": 30188, "known": true, "approximate": false, "prompt": 14351, "completion": 15837, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "medium", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 80.7141, "netPer1kTokens": 1.2505 }, "rank": 15 }, { "modelId": "anthropic/claude-sonnet-5@medium", "baseModelId": "anthropic/claude-sonnet-5", "modelSlug": "claude-sonnet-5", "developerId": "anthropic", "name": "Claude Sonnet 5 (medium)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.4832, "known": true, "approximate": false }, "tokens": { "billed": 59798, "known": true, "approximate": false, "prompt": 14351, "completion": 45447, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "medium", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 78.125, "netPer1kTokens": 0.6313 }, "rank": 16 }, { "modelId": "openai/gpt-5.6-sol@low", "baseModelId": "openai/gpt-5.6-sol", "modelSlug": "gpt-5-6-sol", "developerId": "openai", "name": "GPT-5.6 Sol (low)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.5112, "known": true, "approximate": false }, "tokens": { "billed": 24667, "known": true, "approximate": false, "prompt": 9152, "completion": 15515, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "low", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 73.8459, "netPer1kTokens": 1.5304 }, "rank": 17 }, { "modelId": "x-ai/grok-4.5@medium", "baseModelId": "x-ai/grok-4.5", "modelSlug": "grok-4-5", "developerId": "xai", "name": "Grok 4.5 (medium)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.5223, "known": true, "approximate": false }, "tokens": { "billed": 98216, "known": true, "approximate": false, "prompt": 16749, "completion": 81467, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "medium", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 72.2765, "netPer1kTokens": 0.3844 }, "rank": 18 }, { "modelId": "openai/gpt-5.6-sol@medium", "baseModelId": "openai/gpt-5.6-sol", "modelSlug": "gpt-5-6-sol", "developerId": "openai", "name": "GPT-5.6 Sol (medium)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.5285, "known": true, "approximate": false }, "tokens": { "billed": 25242, "known": true, "approximate": false, "prompt": 9152, "completion": 16090, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "medium", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 71.4286, "netPer1kTokens": 1.4955 }, "rank": 19 }, { "modelId": "anthropic/claude-fable-5@low", "baseModelId": "anthropic/claude-fable-5", "modelSlug": "claude-fable-5", "developerId": "anthropic", "name": "Claude Fable 5 (low)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.5546, "known": true, "approximate": false }, "tokens": { "billed": 22573, "known": true, "approximate": false, "prompt": 14351, "completion": 8222, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "low", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 68.0671, "netPer1kTokens": 1.6724 }, "rank": 20 }, { "modelId": "nvidia/nemotron-3-ultra-550b-a55b@high", "baseModelId": "nvidia/nemotron-3-ultra-550b-a55b", "modelSlug": "nemotron-3-ultra", "developerId": "nvidia", "name": "Nemotron 3 Ultra (high)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.5612, "known": true, "approximate": false }, "tokens": { "billed": 164373, "known": true, "approximate": false, "prompt": 10181, "completion": 154192, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": { "ignore": [ "baseten" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 67.2666, "netPer1kTokens": 0.2297 }, "rank": 21 }, { "modelId": "nvidia/nemotron-3-ultra-550b-a55b@medium", "baseModelId": "nvidia/nemotron-3-ultra-550b-a55b", "modelSlug": "nemotron-3-ultra", "developerId": "nvidia", "name": "Nemotron 3 Ultra (medium)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 0, "invalid": 1, "net": 37.75 } }, "cost": { "usd": 0.5958, "known": true, "approximate": false }, "tokens": { "billed": 173901, "known": true, "approximate": false, "prompt": 10088, "completion": 163813, "orchestration": 0 }, "errors": { "invalid": 1, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "medium", "requestedRoutingPolicy": { "ignore": [ "baseten" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 63.3602, "netPer1kTokens": 0.2171 }, "rank": 22 }, { "modelId": "x-ai/grok-4.5@high", "baseModelId": "x-ai/grok-4.5", "modelSlug": "grok-4-5", "developerId": "xai", "name": "Grok 4.5 (high)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.6047, "known": true, "approximate": false }, "tokens": { "billed": 111954, "known": true, "approximate": false, "prompt": 16749, "completion": 95205, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 62.4277, "netPer1kTokens": 0.3372 }, "rank": 23 }, { "modelId": "google/gemini-3.1-pro-preview@low", "baseModelId": "google/gemini-3.1-pro-preview", "modelSlug": "gemini-3-1-pro", "developerId": "google", "name": "Gemini 3.1 Pro (low)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.6133, "known": true, "approximate": false }, "tokens": { "billed": 58350, "known": true, "approximate": false, "prompt": 8686, "completion": 49664, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "low", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 61.5523, "netPer1kTokens": 0.647 }, "rank": 24 }, { "modelId": "openai/gpt-5.6-sol@high", "baseModelId": "openai/gpt-5.6-sol", "modelSlug": "gpt-5-6-sol", "developerId": "openai", "name": "GPT-5.6 Sol (high)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.6513, "known": true, "approximate": false }, "tokens": { "billed": 29338, "known": true, "approximate": false, "prompt": 9152, "completion": 20186, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 57.961, "netPer1kTokens": 1.2867 }, "rank": 25 }, { "modelId": "qwen/qwen3.7-max", "baseModelId": "qwen/qwen3.7-max", "modelSlug": "qwen3-7-max", "developerId": "alibaba-qwen", "name": "Qwen3.7 Max", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.7187, "known": true, "approximate": false }, "tokens": { "billed": 168909, "known": true, "approximate": false, "prompt": 9734, "completion": 159175, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": null, "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 52.5254, "netPer1kTokens": 0.2235 }, "rank": 26 }, { "modelId": "thinkingmachines/inkling@max", "baseModelId": "thinkingmachines/inkling", "modelSlug": "thinky-inkling", "developerId": "thinking-machines", "name": "Thinky Inkling (max)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.7303, "known": true, "approximate": false }, "tokens": { "billed": 187437, "known": true, "approximate": false, "prompt": 9464, "completion": 177973, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "max", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 51.6911, "netPer1kTokens": 0.2014 }, "rank": 27 }, { "modelId": "openai/gpt-5.6-sol@xhigh", "baseModelId": "openai/gpt-5.6-sol", "modelSlug": "gpt-5-6-sol", "developerId": "openai", "name": "GPT-5.6 Sol (xhigh)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.7315, "known": true, "approximate": false }, "tokens": { "billed": 32011, "known": true, "approximate": false, "prompt": 9152, "completion": 22859, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "xhigh", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 51.6063, "netPer1kTokens": 1.1793 }, "rank": 28 }, { "modelId": "anthropic/claude-opus-5@high", "baseModelId": "anthropic/claude-opus-5", "modelSlug": "claude-opus-5", "developerId": "anthropic", "name": "Claude Opus 5 (high)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.778, "known": true, "approximate": false }, "tokens": { "billed": 42601, "known": true, "approximate": false, "prompt": 14351, "completion": 28250, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 48.5219, "netPer1kTokens": 0.8861 }, "rank": 29 }, { "modelId": "anthropic/claude-sonnet-5@high", "baseModelId": "anthropic/claude-sonnet-5", "modelSlug": "claude-sonnet-5", "developerId": "anthropic", "name": "Claude Sonnet 5 (high)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.8812, "known": true, "approximate": false }, "tokens": { "billed": 99603, "known": true, "approximate": false, "prompt": 14351, "completion": 85252, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 42.8393, "netPer1kTokens": 0.379 }, "rank": 30 }, { "modelId": "google/gemini-3.6-flash@high", "baseModelId": "google/gemini-3.6-flash", "modelSlug": "gemini-3-6-flash", "developerId": "google", "name": "Gemini 3.6 Flash (high)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.9039, "known": true, "approximate": false }, "tokens": { "billed": 127527, "known": true, "approximate": false, "prompt": 8760, "completion": 118767, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 41.7635, "netPer1kTokens": 0.296 }, "rank": 31 }, { "modelId": "google/gemini-3.1-pro-preview@medium", "baseModelId": "google/gemini-3.1-pro-preview", "modelSlug": "gemini-3-1-pro", "developerId": "google", "name": "Gemini 3.1 Pro (medium)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.919, "known": true, "approximate": false }, "tokens": { "billed": 83818, "known": true, "approximate": false, "prompt": 8682, "completion": 75136, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "medium", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 41.0773, "netPer1kTokens": 0.4504 }, "rank": 32 }, { "modelId": "anthropic/claude-opus-5@xhigh", "baseModelId": "anthropic/claude-opus-5", "modelSlug": "claude-opus-5", "developerId": "anthropic", "name": "Claude Opus 5 (xhigh)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.9584, "known": true, "approximate": false }, "tokens": { "billed": 49816, "known": true, "approximate": false, "prompt": 14351, "completion": 35465, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "xhigh", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 39.3886, "netPer1kTokens": 0.7578 }, "rank": 33 }, { "modelId": "anthropic/claude-fable-5@medium", "baseModelId": "anthropic/claude-fable-5", "modelSlug": "claude-fable-5", "developerId": "anthropic", "name": "Claude Fable 5 (medium)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 0.9742, "known": true, "approximate": false }, "tokens": { "billed": 30964, "known": true, "approximate": false, "prompt": 14351, "completion": 16613, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "medium", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 38.7497, "netPer1kTokens": 1.2192 }, "rank": 34 }, { "modelId": "openai/gpt-5.6-sol@max", "baseModelId": "openai/gpt-5.6-sol", "modelSlug": "gpt-5-6-sol", "developerId": "openai", "name": "GPT-5.6 Sol (max)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 1.0607, "known": true, "approximate": false }, "tokens": { "billed": 42982, "known": true, "approximate": false, "prompt": 9152, "completion": 33830, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "max", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 35.5897, "netPer1kTokens": 0.8783 }, "rank": 35 }, { "modelId": "anthropic/claude-sonnet-5@xhigh", "baseModelId": "anthropic/claude-sonnet-5", "modelSlug": "claude-sonnet-5", "developerId": "anthropic", "name": "Claude Sonnet 5 (xhigh)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 1.209, "known": true, "approximate": false }, "tokens": { "billed": 132384, "known": true, "approximate": false, "prompt": 14351, "completion": 118033, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "xhigh", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 31.2242, "netPer1kTokens": 0.2852 }, "rank": 36 }, { "modelId": "anthropic/claude-opus-5@max", "baseModelId": "anthropic/claude-opus-5", "modelSlug": "claude-opus-5", "developerId": "anthropic", "name": "Claude Opus 5 (max)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 1.2753, "known": true, "approximate": false }, "tokens": { "billed": 62493, "known": true, "approximate": false, "prompt": 14351, "completion": 48142, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "max", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 29.6009, "netPer1kTokens": 0.6041 }, "rank": 37 }, { "modelId": "anthropic/claude-fable-5@high", "baseModelId": "anthropic/claude-fable-5", "modelSlug": "claude-fable-5", "developerId": "anthropic", "name": "Claude Fable 5 (high)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 1.4134, "known": true, "approximate": false }, "tokens": { "billed": 39748, "known": true, "approximate": false, "prompt": 14351, "completion": 25397, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 26.7086, "netPer1kTokens": 0.9497 }, "rank": 38 }, { "modelId": "google/gemini-3.1-pro-preview@high", "baseModelId": "google/gemini-3.1-pro-preview", "modelSlug": "gemini-3-1-pro", "developerId": "google", "name": "Gemini 3.1 Pro (high)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 1.4887, "known": true, "approximate": false }, "tokens": { "billed": 131297, "known": true, "approximate": false, "prompt": 8682, "completion": 122615, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 25.3577, "netPer1kTokens": 0.2875 }, "rank": 39 }, { "modelId": "anthropic/claude-fable-5@xhigh", "baseModelId": "anthropic/claude-fable-5", "modelSlug": "claude-fable-5", "developerId": "anthropic", "name": "Claude Fable 5 (xhigh)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 2.0589, "known": true, "approximate": false }, "tokens": { "billed": 52659, "known": true, "approximate": false, "prompt": 14351, "completion": 38308, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "xhigh", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 18.335, "netPer1kTokens": 0.7169 }, "rank": 40 }, { "modelId": "anthropic/claude-fable-5@max", "baseModelId": "anthropic/claude-fable-5", "modelSlug": "claude-fable-5", "developerId": "anthropic", "name": "Claude Fable 5 (max)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 3.3359, "known": true, "approximate": false }, "tokens": { "billed": 78199, "known": true, "approximate": false, "prompt": 14351, "completion": 63848, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "max", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 11.3163, "netPer1kTokens": 0.4827 }, "rank": 41 }, { "modelId": "anthropic/claude-sonnet-5@max", "baseModelId": "anthropic/claude-sonnet-5", "modelSlug": "claude-sonnet-5", "developerId": "anthropic", "name": "Claude Sonnet 5 (max)", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": 4.0601, "known": true, "approximate": false }, "tokens": { "billed": 417489, "known": true, "approximate": false, "prompt": 14351, "completion": 403138, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "max", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 9.2978, "netPer1kTokens": 0.0904 }, "rank": 42 }, { "modelId": "openrouter/fusion", "baseModelId": "openrouter/fusion", "modelSlug": "openrouter-fusion", "developerId": "openrouter", "name": "OpenRouter Fusion", "score": 213.25, "net": 37.75, "sections": { "MAT": { "correct": 38, "wrong": 1, "invalid": 0, "net": 37.75 } }, "cost": { "usd": null, "known": false, "approximate": false }, "tokens": { "billed": 94666, "known": true, "approximate": false, "prompt": 57570, "completion": 37096, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": null, "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": null, "netPer1kTokens": 0.3988 }, "rank": 43 }, { "modelId": "openai/gpt-5.6-luna@medium", "baseModelId": "openai/gpt-5.6-luna", "modelSlug": "gpt-5-6-luna", "developerId": "openai", "name": "GPT-5.6 Luna (medium)", "score": 209.5, "net": 36.5, "sections": { "MAT": { "correct": 37, "wrong": 2, "invalid": 0, "net": 36.5 } }, "cost": { "usd": 0.0154, "known": true, "approximate": false }, "tokens": { "billed": 33326, "known": true, "approximate": false, "prompt": 9152, "completion": 24174, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "medium", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 2370.1299, "netPer1kTokens": 1.0952 }, "rank": 44 }, { "modelId": "tencent/hy3@low", "baseModelId": "tencent/hy3", "modelSlug": "tencent-hy3", "developerId": "tencent", "name": "Tencent Hy3 (low)", "score": 209.5, "net": 36.5, "sections": { "MAT": { "correct": 37, "wrong": 2, "invalid": 0, "net": 36.5 } }, "cost": { "usd": 0.0479, "known": true, "approximate": true }, "tokens": { "billed": 98707, "known": true, "approximate": false, "prompt": 10756, "completion": 87951, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": { "tr": "Maliyet liste fiyatından hesaplandı; allow_fallbacks nedeniyle 2026-07-31 ölçümünde gerçek fatura hesaplananın ≈1,5 katıydı.", "en": "Cost computed from the list price; with allow_fallbacks, a 2026-07-31 probe measured the actual bill at ≈1.5× the computed value." }, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "low", "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 762.0042, "netPer1kTokens": 0.3698 }, "rank": 45 }, { "modelId": "thinkingmachines/inkling@minimal", "baseModelId": "thinkingmachines/inkling", "modelSlug": "thinky-inkling", "developerId": "thinking-machines", "name": "Thinky Inkling (minimal)", "score": 209.5, "net": 36.5, "sections": { "MAT": { "correct": 37, "wrong": 2, "invalid": 0, "net": 36.5 } }, "cost": { "usd": 0.0887, "known": true, "approximate": false }, "tokens": { "billed": 29026, "known": true, "approximate": false, "prompt": 9464, "completion": 19562, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "minimal", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 411.4994, "netPer1kTokens": 1.2575 }, "rank": 46 }, { "modelId": "xiaomi/mimo-v2.5-pro", "baseModelId": "xiaomi/mimo-v2.5-pro", "modelSlug": "mimo-v2-5-pro", "developerId": "xiaomi", "name": "MiMo V2.5 Pro", "score": 209.5, "net": 36.5, "sections": { "MAT": { "correct": 37, "wrong": 2, "invalid": 0, "net": 36.5 } }, "cost": { "usd": 0.097, "known": true, "approximate": false }, "tokens": { "billed": 116515, "known": true, "approximate": false, "prompt": 10096, "completion": 106419, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": null, "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 376.2887, "netPer1kTokens": 0.3133 }, "rank": 47 }, { "modelId": "deepseek/deepseek-v4-pro@high", "baseModelId": "deepseek/deepseek-v4-pro", "modelSlug": "deepseek-v4-pro", "developerId": "deepseek", "name": "DeepSeek V4 Pro (high)", "score": 209.5, "net": 36.5, "sections": { "MAT": { "correct": 37, "wrong": 1, "invalid": 1, "net": 36.5 } }, "cost": { "usd": 0.1133, "known": true, "approximate": true }, "tokens": { "billed": 135789, "known": true, "approximate": false, "prompt": 11228, "completion": 124561, "orchestration": 0 }, "errors": { "invalid": 1, "apiFailures": 0 }, "note": { "tr": "Maliyet liste fiyatından hesaplandı; model allow_fallbacks ile koşuldu ve 2026-07-31 usage.cost ölçümünde gerçek fatura hesaplananın ≈3,9 katına ulaştı.", "en": "Cost computed from the list price; the model ran with allow_fallbacks and a 2026-07-31 usage.cost probe measured the actual bill at ≈3.9× the computed value." }, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 322.1536, "netPer1kTokens": 0.2688 }, "rank": 48 }, { "modelId": "openai/gpt-5.6-terra@medium", "baseModelId": "openai/gpt-5.6-terra", "modelSlug": "gpt-5-6-terra", "developerId": "openai", "name": "GPT-5.6 Terra (medium)", "score": 209.5, "net": 36.5, "sections": { "MAT": { "correct": 37, "wrong": 2, "invalid": 0, "net": 36.5 } }, "cost": { "usd": 0.1357, "known": true, "approximate": false }, "tokens": { "billed": 30237, "known": true, "approximate": false, "prompt": 9152, "completion": 21085, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "medium", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 268.9757, "netPer1kTokens": 1.2071 }, "rank": 49 }, { "modelId": "deepseek/deepseek-v4-flash@xhigh", "baseModelId": "deepseek/deepseek-v4-flash", "modelSlug": "deepseek-v4-flash", "developerId": "deepseek", "name": "DeepSeek V4 Flash (xhigh)", "score": 209.5, "net": 36.5, "sections": { "MAT": { "correct": 37, "wrong": 0, "invalid": 2, "net": 36.5 } }, "cost": { "usd": 0.1532, "known": true, "approximate": false }, "tokens": { "billed": 554206, "known": true, "approximate": false, "prompt": 14452, "completion": 539754, "orchestration": 0 }, "errors": { "invalid": 2, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "xhigh", "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "ignore": [ "akashml", "io-net" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 238.2507, "netPer1kTokens": 0.0659 }, "rank": 50 }, { "modelId": "google/gemini-3.6-flash@minimal", "baseModelId": "google/gemini-3.6-flash", "modelSlug": "gemini-3-6-flash", "developerId": "google", "name": "Gemini 3.6 Flash (minimal)", "score": 209.5, "net": 36.5, "sections": { "MAT": { "correct": 37, "wrong": 2, "invalid": 0, "net": 36.5 } }, "cost": { "usd": 0.2213, "known": true, "approximate": false }, "tokens": { "billed": 36516, "known": true, "approximate": false, "prompt": 8760, "completion": 27756, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "minimal", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 164.9345, "netPer1kTokens": 0.9996 }, "rank": 51 }, { "modelId": "anthropic/claude-sonnet-5@low", "baseModelId": "anthropic/claude-sonnet-5", "modelSlug": "claude-sonnet-5", "developerId": "anthropic", "name": "Claude Sonnet 5 (low)", "score": 209.5, "net": 36.5, "sections": { "MAT": { "correct": 37, "wrong": 2, "invalid": 0, "net": 36.5 } }, "cost": { "usd": 0.2561, "known": true, "approximate": false }, "tokens": { "billed": 37093, "known": true, "approximate": false, "prompt": 14351, "completion": 22742, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "low", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 142.5225, "netPer1kTokens": 0.984 }, "rank": 52 }, { "modelId": "x-ai/grok-4.5@low", "baseModelId": "x-ai/grok-4.5", "modelSlug": "grok-4-5", "developerId": "xai", "name": "Grok 4.5 (low)", "score": 209.5, "net": 36.5, "sections": { "MAT": { "correct": 37, "wrong": 1, "invalid": 1, "net": 36.5 } }, "cost": { "usd": 0.3234, "known": true, "approximate": false }, "tokens": { "billed": 65065, "known": true, "approximate": false, "prompt": 16749, "completion": 48316, "orchestration": 0 }, "errors": { "invalid": 1, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "low", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 112.8633, "netPer1kTokens": 0.561 }, "rank": 53 }, { "modelId": "thinkingmachines/inkling@high", "baseModelId": "thinkingmachines/inkling", "modelSlug": "thinky-inkling", "developerId": "thinking-machines", "name": "Thinky Inkling (high)", "score": 209.5, "net": 36.5, "sections": { "MAT": { "correct": 37, "wrong": 1, "invalid": 1, "net": 36.5 } }, "cost": { "usd": 0.4184, "known": true, "approximate": false }, "tokens": { "billed": 110403, "known": true, "approximate": false, "prompt": 9411, "completion": 100992, "orchestration": 0 }, "errors": { "invalid": 1, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 87.2371, "netPer1kTokens": 0.3306 }, "rank": 54 }, { "modelId": "meituan/longcat-2.0", "baseModelId": "meituan/longcat-2.0", "modelSlug": "longcat-2-0", "developerId": "meituan", "name": "LongCat 2.0", "score": 209.5, "net": 36.5, "sections": { "MAT": { "correct": 37, "wrong": 2, "invalid": 0, "net": 36.5 } }, "cost": { "usd": 0.4344, "known": true, "approximate": false }, "tokens": { "billed": 370865, "known": true, "approximate": false, "prompt": 11845, "completion": 359020, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": null, "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 84.0239, "netPer1kTokens": 0.0984 }, "rank": 55 }, { "modelId": "moonshotai/kimi-k3@low", "baseModelId": "moonshotai/kimi-k3", "modelSlug": "kimi-k3", "developerId": "moonshot", "name": "Kimi K3 (low)", "score": 209.5, "net": 36.5, "sections": { "MAT": { "correct": 37, "wrong": 1, "invalid": 1, "net": 36.5 } }, "cost": { "usd": 0.5678, "known": true, "approximate": false }, "tokens": { "billed": 49585, "known": true, "approximate": false, "prompt": 14662, "completion": 34923, "orchestration": 0 }, "errors": { "invalid": 1, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "low", "requestedRoutingPolicy": { "ignore": [ "nebius", "moonshotai" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 64.2832, "netPer1kTokens": 0.7361 }, "rank": 56 }, { "modelId": "moonshotai/kimi-k3@high", "baseModelId": "moonshotai/kimi-k3", "modelSlug": "kimi-k3", "developerId": "moonshot", "name": "Kimi K3 (high)", "score": 209.5, "net": 36.5, "sections": { "MAT": { "correct": 37, "wrong": 1, "invalid": 1, "net": 36.5 } }, "cost": { "usd": 0.7433, "known": true, "approximate": false }, "tokens": { "billed": 61432, "known": true, "approximate": false, "prompt": 14850, "completion": 46582, "orchestration": 0 }, "errors": { "invalid": 1, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": { "ignore": [ "nebius", "moonshotai" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 49.1053, "netPer1kTokens": 0.5942 }, "rank": 57 }, { "modelId": "google/gemini-3.6-flash@medium", "baseModelId": "google/gemini-3.6-flash", "modelSlug": "gemini-3-6-flash", "developerId": "google", "name": "Gemini 3.6 Flash (medium)", "score": 209.5, "net": 36.5, "sections": { "MAT": { "correct": 37, "wrong": 2, "invalid": 0, "net": 36.5 } }, "cost": { "usd": 0.797, "known": true, "approximate": false }, "tokens": { "billed": 113281, "known": true, "approximate": false, "prompt": 8760, "completion": 104521, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "medium", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 45.7967, "netPer1kTokens": 0.3222 }, "rank": 58 }, { "modelId": "google/gemma-4-31b-it", "baseModelId": "google/gemma-4-31b-it", "modelSlug": "gemma-4-31b", "developerId": "google", "name": "Gemma 4 31B", "score": 205.75, "net": 35.25, "sections": { "MAT": { "correct": 36, "wrong": 3, "invalid": 0, "net": 35.25 } }, "cost": { "usd": 0.0131, "known": true, "approximate": true }, "tokens": { "billed": 45262, "known": true, "approximate": false, "prompt": 9385, "completion": 35877, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": { "tr": "Maliyet liste fiyatından hesaplandı; allow_fallbacks nedeniyle 2026-07-31 ölçümünde gerçek fatura hesaplananın ≈1,2 katıydı.", "en": "Cost computed from the list price; with allow_fallbacks, a 2026-07-31 probe measured the actual bill at ≈1.2× the computed value." }, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": null, "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 2690.8397, "netPer1kTokens": 0.7788 }, "rank": 59 }, { "modelId": "deepseek/deepseek-v4-pro@xhigh", "baseModelId": "deepseek/deepseek-v4-pro", "modelSlug": "deepseek-v4-pro", "developerId": "deepseek", "name": "DeepSeek V4 Pro (xhigh)", "score": 205.75, "net": 35.25, "sections": { "MAT": { "correct": 36, "wrong": 0, "invalid": 3, "net": 35.25 } }, "cost": { "usd": 0.2256, "known": true, "approximate": true }, "tokens": { "billed": 266260, "known": true, "approximate": false, "prompt": 14012, "completion": 252248, "orchestration": 0 }, "errors": { "invalid": 3, "apiFailures": 0 }, "note": { "tr": "Maliyet liste fiyatından hesaplandı; model allow_fallbacks ile koşuldu ve 2026-07-31 usage.cost ölçümünde gerçek fatura hesaplananın ≈3,9 katına ulaştı.", "en": "Cost computed from the list price; the model ran with allow_fallbacks and a 2026-07-31 usage.cost probe measured the actual bill at ≈3.9× the computed value." }, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "xhigh", "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 156.25, "netPer1kTokens": 0.1324 }, "rank": 60 }, { "modelId": "z-ai/glm-5.2@xhigh", "baseModelId": "z-ai/glm-5.2", "modelSlug": "glm-5-2", "developerId": "zai", "name": "GLM 5.2 (xhigh)", "score": 205.75, "net": 35.25, "sections": { "MAT": { "correct": 36, "wrong": 1, "invalid": 2, "net": 35.25 } }, "cost": { "usd": 1.4658, "known": true, "approximate": true }, "tokens": { "billed": 398849, "known": true, "approximate": false, "prompt": 10155, "completion": 388694, "orchestration": 0 }, "errors": { "invalid": 2, "apiFailures": 0 }, "note": { "tr": "Maliyet liste fiyatından hesaplandı; allow_fallbacks nedeniyle 2026-07-31 ölçümünde gerçek fatura hesaplananın ≈1,3 katıydı.", "en": "Cost computed from the list price; with allow_fallbacks, a 2026-07-31 probe measured the actual bill at ≈1.3× the computed value." }, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "xhigh", "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "ignore": [ "akashml", "ambient" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 24.0483, "netPer1kTokens": 0.0884 }, "rank": 61 }, { "modelId": "moonshotai/kimi-k3@max", "baseModelId": "moonshotai/kimi-k3", "modelSlug": "kimi-k3", "developerId": "moonshot", "name": "Kimi K3 (max)", "score": 205.75, "net": 35.25, "sections": { "MAT": { "correct": 36, "wrong": 1, "invalid": 2, "net": 35.25 } }, "cost": { "usd": 3.1251, "known": true, "approximate": false }, "tokens": { "billed": 220059, "known": true, "approximate": false, "prompt": 14647, "completion": 205412, "orchestration": 0 }, "errors": { "invalid": 2, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "max", "requestedRoutingPolicy": { "ignore": [ "nebius", "moonshotai" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 11.2796, "netPer1kTokens": 0.1602 }, "rank": 62 }, { "modelId": "openai/gpt-5.6-luna@high", "baseModelId": "openai/gpt-5.6-luna", "modelSlug": "gpt-5-6-luna", "developerId": "openai", "name": "GPT-5.6 Luna (high)", "score": 202.0, "net": 34.0, "sections": { "MAT": { "correct": 35, "wrong": 4, "invalid": 0, "net": 34.0 } }, "cost": { "usd": 0.0258, "known": true, "approximate": false }, "tokens": { "billed": 50682, "known": true, "approximate": false, "prompt": 9152, "completion": 41530, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 1317.8295, "netPer1kTokens": 0.6708 }, "rank": 63 }, { "modelId": "thinkingmachines/inkling@low", "baseModelId": "thinkingmachines/inkling", "modelSlug": "thinky-inkling", "developerId": "thinking-machines", "name": "Thinky Inkling (low)", "score": 202.0, "net": 34.0, "sections": { "MAT": { "correct": 35, "wrong": 4, "invalid": 0, "net": 34.0 } }, "cost": { "usd": 0.0762, "known": true, "approximate": false }, "tokens": { "billed": 25949, "known": true, "approximate": false, "prompt": 9464, "completion": 16485, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "low", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 446.1942, "netPer1kTokens": 1.3103 }, "rank": 64 }, { "modelId": "deepseek/deepseek-v4-flash@high", "baseModelId": "deepseek/deepseek-v4-flash", "modelSlug": "deepseek-v4-flash", "developerId": "deepseek", "name": "DeepSeek V4 Flash (high)", "score": 202.0, "net": 34.0, "sections": { "MAT": { "correct": 35, "wrong": 1, "invalid": 3, "net": 34.0 } }, "cost": { "usd": 0.1483, "known": true, "approximate": false }, "tokens": { "billed": 535441, "known": true, "approximate": false, "prompt": 11320, "completion": 524121, "orchestration": 0 }, "errors": { "invalid": 3, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "ignore": [ "akashml", "io-net" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 229.265, "netPer1kTokens": 0.0635 }, "rank": 65 }, { "modelId": "openai/gpt-5.6-sol@none", "baseModelId": "openai/gpt-5.6-sol", "modelSlug": "gpt-5-6-sol", "developerId": "openai", "name": "GPT-5.6 Sol (none)", "score": 202.0, "net": 34.0, "sections": { "MAT": { "correct": 35, "wrong": 4, "invalid": 0, "net": 34.0 } }, "cost": { "usd": 0.3261, "known": true, "approximate": false }, "tokens": { "billed": 18497, "known": true, "approximate": false, "prompt": 9152, "completion": 9345, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "none", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 104.2625, "netPer1kTokens": 1.8381 }, "rank": 66 }, { "modelId": "minimax/minimax-m3", "baseModelId": "minimax/minimax-m3", "modelSlug": "minimax-m3", "developerId": "minimax", "name": "MiniMax M3", "score": 202.0, "net": 34.0, "sections": { "MAT": { "correct": 35, "wrong": 1, "invalid": 3, "net": 34.0 } }, "cost": { "usd": 0.5905, "known": true, "approximate": false }, "tokens": { "billed": 505117, "known": true, "approximate": false, "prompt": 17428, "completion": 487689, "orchestration": 0 }, "errors": { "invalid": 3, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": null, "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 57.5783, "netPer1kTokens": 0.0673 }, "rank": 67 }, { "modelId": "qwen/qwen3.6-27b", "baseModelId": "qwen/qwen3.6-27b", "modelSlug": "qwen3-6-27b", "developerId": "alibaba-qwen", "name": "Qwen3.6 27B", "score": 202.0, "net": 34.0, "sections": { "MAT": { "correct": 35, "wrong": 2, "invalid": 2, "net": 34.0 } }, "cost": { "usd": 0.6567, "known": true, "approximate": true }, "tokens": { "billed": 336631, "known": true, "approximate": false, "prompt": 9734, "completion": 326897, "orchestration": 0 }, "errors": { "invalid": 2, "apiFailures": 0 }, "note": { "tr": "Maliyet liste fiyatından hesaplandı; allow_fallbacks nedeniyle 2026-07-31 ölçümünde gerçek fatura hesaplananın ≈1,8 katıydı.", "en": "Cost computed from the list price; with allow_fallbacks, a 2026-07-31 probe measured the actual bill at ≈1.8× the computed value." }, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": null, "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "ignore": [ "io-net" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 51.774, "netPer1kTokens": 0.101 }, "rank": 68 }, { "modelId": "z-ai/glm-5.2@high", "baseModelId": "z-ai/glm-5.2", "modelSlug": "glm-5-2", "developerId": "zai", "name": "GLM 5.2 (high)", "score": 202.0, "net": 34.0, "sections": { "MAT": { "correct": 35, "wrong": 2, "invalid": 2, "net": 34.0 } }, "cost": { "usd": 1.2107, "known": true, "approximate": true }, "tokens": { "billed": 330631, "known": true, "approximate": false, "prompt": 10153, "completion": 320478, "orchestration": 0 }, "errors": { "invalid": 2, "apiFailures": 0 }, "note": { "tr": "Maliyet liste fiyatından hesaplandı; allow_fallbacks nedeniyle 2026-07-31 ölçümünde gerçek fatura hesaplananın ≈1,3 katıydı.", "en": "Cost computed from the list price; with allow_fallbacks, a 2026-07-31 probe measured the actual bill at ≈1.3× the computed value." }, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "high", "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "ignore": [ "akashml", "ambient" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 28.0829, "netPer1kTokens": 0.1028 }, "rank": 69 }, { "modelId": "tencent/hy3@none", "baseModelId": "tencent/hy3", "modelSlug": "tencent-hy3", "developerId": "tencent", "name": "Tencent Hy3 (none)", "score": 198.25, "net": 32.75, "sections": { "MAT": { "correct": 34, "wrong": 5, "invalid": 0, "net": 32.75 } }, "cost": { "usd": 0.0124, "known": true, "approximate": true }, "tokens": { "billed": 31729, "known": true, "approximate": false, "prompt": 10873, "completion": 20856, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": { "tr": "Maliyet liste fiyatından hesaplandı; allow_fallbacks nedeniyle 2026-07-31 ölçümünde gerçek fatura hesaplananın ≈1,5 katıydı.", "en": "Cost computed from the list price; with allow_fallbacks, a 2026-07-31 probe measured the actual bill at ≈1.5× the computed value." }, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "none", "requestedRoutingPolicy": { "quantizations": [ "fp8" ], "allow_fallbacks": true }, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 2641.129, "netPer1kTokens": 1.0322 }, "rank": 70 }, { "modelId": "openai/gpt-5.6-luna@low", "baseModelId": "openai/gpt-5.6-luna", "modelSlug": "gpt-5-6-luna", "developerId": "openai", "name": "GPT-5.6 Luna (low)", "score": 198.25, "net": 32.75, "sections": { "MAT": { "correct": 34, "wrong": 5, "invalid": 0, "net": 32.75 } }, "cost": { "usd": 0.0165, "known": true, "approximate": false }, "tokens": { "billed": 35050, "known": true, "approximate": false, "prompt": 9152, "completion": 25898, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "low", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 1984.8485, "netPer1kTokens": 0.9344 }, "rank": 71 }, { "modelId": "thinkingmachines/inkling@none", "baseModelId": "thinkingmachines/inkling", "modelSlug": "thinky-inkling", "developerId": "thinking-machines", "name": "Thinky Inkling (none)", "score": 187.0, "net": 29.0, "sections": { "MAT": { "correct": 31, "wrong": 8, "invalid": 0, "net": 29.0 } }, "cost": { "usd": 0.1154, "known": true, "approximate": false }, "tokens": { "billed": 35568, "known": true, "approximate": false, "prompt": 9394, "completion": 26174, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "none", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 251.2998, "netPer1kTokens": 0.8153 }, "rank": 72 }, { "modelId": "openai/gpt-5.6-luna@none", "baseModelId": "openai/gpt-5.6-luna", "modelSlug": "gpt-5-6-luna", "developerId": "openai", "name": "GPT-5.6 Luna (none)", "score": 179.5, "net": 26.5, "sections": { "MAT": { "correct": 29, "wrong": 10, "invalid": 0, "net": 26.5 } }, "cost": { "usd": 0.0074, "known": true, "approximate": false }, "tokens": { "billed": 20011, "known": true, "approximate": false, "prompt": 9152, "completion": 10859, "orchestration": 0 }, "errors": { "invalid": 0, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "none", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 3581.0811, "netPer1kTokens": 1.3243 }, "rank": 73 }, { "modelId": "openai/gpt-5.6-terra@none", "baseModelId": "openai/gpt-5.6-terra", "modelSlug": "gpt-5-6-terra", "developerId": "openai", "name": "GPT-5.6 Terra (none)", "score": 179.5, "net": 26.5, "sections": { "MAT": { "correct": 29, "wrong": 9, "invalid": 1, "net": 26.5 } }, "cost": { "usd": 0.0563, "known": true, "approximate": false }, "tokens": { "billed": 17009, "known": true, "approximate": false, "prompt": 9152, "completion": 7857, "orchestration": 0 }, "errors": { "invalid": 1, "apiFailures": 0 }, "note": null, "settings": { "temperature": 0.0, "maxTokens": 131072, "reasoningEffort": "none", "requestedRoutingPolicy": null, "endpointBaseUrl": null, "resolvedInferenceProviderId": null, "resolutionSource": "unknown", "settingsSource": "run" }, "efficiency": { "netPerUsd": 470.6927, "netPer1kTokens": 1.558 }, "rank": 74 } ], "questionStats": [ { "id": "MAT-1", "number": 1, "section": "MAT", "cancelled": false, "correct": 72, "responses": 74 }, { "id": "MAT-2", "number": 2, "section": "MAT", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "MAT-3", "number": 3, "section": "MAT", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "MAT-4", "number": 4, "section": "MAT", "cancelled": false, "correct": 61, "responses": 74 }, { "id": "MAT-5", "number": 5, "section": "MAT", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "MAT-6", "number": 6, "section": "MAT", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "MAT-7", "number": 7, "section": "MAT", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "MAT-8", "number": 8, "section": "MAT", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "MAT-9", "number": 9, "section": "MAT", "cancelled": false, "correct": 0, "responses": 74 }, { "id": "MAT-10", "number": 10, "section": "MAT", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "MAT-11", "number": 11, "section": "MAT", "cancelled": false, "correct": 72, "responses": 74 }, { "id": "MAT-12", "number": 12, "section": "MAT", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "MAT-13", "number": 13, "section": "MAT", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "MAT-14", "number": 14, "section": "MAT", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "MAT-15", "number": 15, "section": "MAT", "cancelled": false, "correct": 67, "responses": 74 }, { "id": "MAT-16", "number": 16, "section": "MAT", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "MAT-17", "number": 17, "section": "MAT", "cancelled": false, "correct": 72, "responses": 74 }, { "id": "MAT-18", "number": 18, "section": "MAT", "cancelled": false, "correct": 72, "responses": 74 }, { "id": "MAT-19", "number": 19, "section": "MAT", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "MAT-20", "number": 20, "section": "MAT", "cancelled": true, "correct": null, "responses": null }, { "id": "MAT-21", "number": 21, "section": "MAT", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "MAT-22", "number": 22, "section": "MAT", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "MAT-23", "number": 23, "section": "MAT", "cancelled": false, "correct": 68, "responses": 74 }, { "id": "MAT-24", "number": 24, "section": "MAT", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "MAT-25", "number": 25, "section": "MAT", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "MAT-26", "number": 26, "section": "MAT", "cancelled": false, "correct": 71, "responses": 74 }, { "id": "MAT-27", "number": 27, "section": "MAT", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "MAT-28", "number": 28, "section": "MAT", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "MAT-29", "number": 29, "section": "MAT", "cancelled": false, "correct": 72, "responses": 74 }, { "id": "MAT-30", "number": 30, "section": "MAT", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "MAT-31", "number": 31, "section": "MAT", "cancelled": false, "correct": 73, "responses": 74 }, { "id": "MAT-32", "number": 32, "section": "MAT", "cancelled": false, "correct": 67, "responses": 74 }, { "id": "MAT-33", "number": 33, "section": "MAT", "cancelled": false, "correct": 71, "responses": 74 }, { "id": "MAT-34", "number": 34, "section": "MAT", "cancelled": false, "correct": 68, "responses": 74 }, { "id": "MAT-35", "number": 35, "section": "MAT", "cancelled": false, "correct": 68, "responses": 74 }, { "id": "MAT-36", "number": 36, "section": "MAT", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "MAT-37", "number": 37, "section": "MAT", "cancelled": false, "correct": 66, "responses": 74 }, { "id": "MAT-38", "number": 38, "section": "MAT", "cancelled": false, "correct": 72, "responses": 74 }, { "id": "MAT-39", "number": 39, "section": "MAT", "cancelled": false, "correct": 74, "responses": 74 }, { "id": "MAT-40", "number": 40, "section": "MAT", "cancelled": false, "correct": 73, "responses": 74 } ] }